[project]Port android to meizu M8

MartO,
sorry im not beeing able to provide much help for the moment but i am running out of time :s

but i did look at the basic logs from haret, and i am surprised that it still does not support the cpu and the log are allmost the same :s
However you have made a new step forward, due the fact that compiling haret worked;-)

Did i understand you correctly ;
you have build the android kernel with M8 cpu support but need the other files? so yes, this means you have the zimage ?
In this case try Kaiser Builds [Android on HTC]
from here you will gain an example on the default file needed for haret,
initrd and system image aswel.

The default.txt will need to be edited to your kernel, but then you can give it a try, just to see what it gives.

Ill make time this sunday to start compiling android itself,
then we can provide a download link for this.

LVL and MaikelRunia, how more documentation we get the better it becomes , and thanks helping out ;-)

Go on guys !!Big up !!
We follow you !!

A little update:

Compiling this Android kernel:
git://github.com/iamnpc/samsung-s3c6410-android.1.0.git

In Mint Linux, following this guide:
Android Updates

I get this error:

Code:

arch/arm/mach-s3c6410/built-in.o: In function `smdk6410_machine_init':
pm.c:(.init.text+0x5e8): undefined reference to `i2c_register_board_info'
pm.c:(.init.text+0x5f8): undefined reference to `i2c_register_board_info'
kernel/built-in.o: In function `timekeeping_resume':
tsacct.c:(.text+0x200a8): undefined reference to `__umoddi3'
tsacct.c:(.text+0x200cc): undefined reference to `__udivdi3'
kernel/built-in.o: In function `update_wall_time':
tsacct.c:(.text+0x206b4): undefined reference to `__umoddi3'
tsacct.c:(.text+0x206d4): undefined reference to `__udivdi3'
make: *** [.tmp_vmlinux1] Error

I tried to modify the Kernel config using "make menuconfig" but still error's.
Also I tried "-i" option to "discard errors" and I get this error:

Code:

arch/arm/mach-s3c6410/built-in.o: In function `smdk6410_machine_init':
pm.c:(.init.text+0x5e8): undefined reference to `i2c_register_board_info'
pm.c:(.init.text+0x5f8): undefined reference to `i2c_register_board_info'
kernel/built-in.o: In function `timekeeping_resume':
tsacct.c:(.text+0x200a8): undefined reference to `__umoddi3'
tsacct.c:(.text+0x200cc): undefined reference to `__udivdi3'
kernel/built-in.o: In function `update_wall_time':
tsacct.c:(.text+0x206b4): undefined reference to `__umoddi3'
tsacct.c:(.text+0x206d4): undefined reference to `__udivdi3'
make: [vmlinux] Error 1 (ignored)
  SYSMAP  System.map
arm-none-linux-gnueabi-nm: 'vmlinux': No such file
make: [vmlinux] Error 1 (ignored)
  SYSMAP  .tmp_System.map
arm-none-linux-gnueabi-nm: '.tmp_vmlinux2': No such file
make: [vmlinux] Error 1 (ignored)
make[1]: *** No rule to make target `vmlinux', needed by `arch/arm/boot/Image'.  Stop.
make: [zImage] Error 2 (ignored)

Best Regards

MartO,
did you just get the git file and went straight to the compile or?

You see... I think that was my problem...

I got the GIT and then runned "make menuconfig" to configure the kernel and I changed lot's of stuff like taking out drivers... maybe I broke something.

After a good sleep... I'll download the GIT again and go straight to compile it to see what happen... But I think that I need a ".config" file to compile it...

UPDATE: SAME PROBLEM... hitting a wall now.

that could be true,
try just to compile one to see what it does without making changes?

if oneday you really would like to compile android from scratch i would use the original git from google.

Get a good rest and try it again ;-)
sleep tight

ps i did see your update to late : get some rest first

Android from scratch I would be great...

Actually the other day I compiled Android Source and It was fairly easy (It took about 6hrs!!).

The problem with the Source is the lack of support o M8's CPU...

It would be real nice to have some Meizu engineer support on this one.

however this is a low priority project for me i was browsing the web,
and found maybe a new lead, not sure if it will work but this is the explanation in raw lines:

*place haret on device
*point in default.txt to load Uboot
*place android kernel on disk
*point Uboot to android
the idea comes from :
HpIpaq214HowTo - Handhelds.org MoinMoin Wiki

There is a remark ive read :
it is mentioned on the above link aswel :

Booting from HaRET starts linux with all the hardware in whatever unpredictable state winCE left it in. This usually causes problems of various sorts including that the USB Host system will not work at all.

SO maybe, maybe we need a bootloader that removes wince complete.
but that will run the device un-usable until android is finished :/

So a question to you folks, if we would build android that supports m8 can it allready been tested before porting it? Meaning emulate it in windows ,...?
Does anyone knows the answer?

MartO,
i fear that meizu will not support our work
it would be great if they did ofcourse:-).

Now the cpu support is not really work for meizu, thats samsungs job.
did you allready have a look here?
About Google Android
android-group-korea - Google Code

ps i gues that you are becoming good in it? ;-)


update:
next to haret and uboot as bootloader there seems also on called Qi:
source : http://wiki.openmoko.org/wiki/Qi
there seems support for the m8 cpu (see in list):
http://people.openmoko.org/andy/

*Updated howto parts with compile info on haret.



Can someone compile Qi source with s3c6410 support?

then run it following this way:
*install haret with s3c6410 support
*edit default.txt and point it to location Qi bootloader
*install Qi with s3c6410 support
*backup and restart m8 device
*run haret
*select run to run the default.txt file
*provide feedback what it does?

=> it should restart system and boot qi, cause qi has no linux install it should only perform a memory test.

=> if qi boots it will look on all storage paths for a kernel to boot,
so you could try to place a htc kernel just to see if it would try to mount..

More info on OpenMoko's Qi Boot-loader:

Qi - Openmoko

To get the source code:

In a Linux machine:
1-. go to Terminal
2-. go to you chosen folder (where you want to put the source code).
3-. and then: "git clone git://git.openmoko.org/git/qi.git" (without the "").

It is also very important to get the Cross Compiler to compile the file... in this case it's "arm-angstrom-linux-gnueabi".

GO HERE for the Toolchain / Compiler
Toolchain - Openmoko

Now... where is the S3C6410 patch? UPDATE --> no patch necessary as it supports S3C6410...

But yet again I'm having problem with compiling!!!

I'm pretty sure I'm doing everything OK. I install the compiler in the correct directory and then add the correct path (see this page on how to add the path it's very good:
HOWTO: Add a Directory to my Path Statement/Variable. : New Linux User - Linux Tips - Information for Using Linux )

I get this error :
marto@marto-laptop:~/qi$ make cpu=s3c6410
/usr/local/openmoko/arm/bin/arm-angstrom-linux-gnueabi-gcc: 1: Syntax error: "(" unexpected
make: *** [src/blink_led.o] Error 2
marto@marto-laptop:~/qi$

UPDATE --> I think I downloaded the 64bit version of the compiler. Me = dumbass.

Originally Posted by evow04

*Updated howto parts with compile info on haret.



Can someone compile Qi source with s3c6410 support?

then run it following this way:
*install haret with s3c6410 support
*edit default.txt and point it to location Qi bootloader
*install Qi with s3c6410 support
*backup and restart m8 device
*run haret
*select run to run the default.txt file
*provide feedback what it does?

=> it should restart system and boot qi, cause qi has no linux install it should only perform a memory test.

=> if qi boots it will look on all storage paths for a kernel to boot,
so you could try to place a htc kernel just to see if it would try to mount..

Ok, got to compile Qi with S3C6410 support. It generates a file called:
"start_qi_all-s3c6410"

Code:

marto@marto-laptop:~/qi$ make CPU=s3c6410
mkdir -p image
===> DFU Trailer information:
Trailer Vers.:    1
Trailer Length:    16
VendorID:    0x1d50
ProductID:    0x5119
HW Revision:    0x0350

Now...
In Haret's default.txt should I point this file as kernel? initrd? or what?

I go to sleep now...

Hi there MartO,
i am always glad to hear you :-)

the file "start_qi_all-s3c6410", is this a xxxx.bin file?

Onto your questions:
according the haret documentation the following is necessarily in the default.txt:

Code:

set kernel <full_filename>
boot

but there are options aswel :

Code:

set initrd <filename>
set mtype <mtype>
ramaddr
ramsize

Those options are not necessarily for the moment but could/will be in the future.
Ram size + address could be obtained by just running haret and look into the logs what it detects. what sound obvious.

The mtype stands for "Linux kernel machine type" , according this page ARM Linux - Developer - Machines it should be

Code:

Type ID:     1626

So here is an example with the above information mentioned,
please remind i can not test it myself cause i still dont have my m8!
default.txt :

Code:

set MTYPE     1626
set KERNEL start_qi_all-s3c6410
boot

ps:
I would like to mention that if you are looking on info for the m8, google words like : s3c6410(cpu samsung) , smdk6410(testboard samsung cpu)

ps 2 :
if you managed to compile qi i will add the howto from your previous post onto the first page ok?

hey... I only wanted to thank you for investing so much time
I hope you will get it!

In a few days I have holidays and will stay @home, maybe I can help a little bit!

Hi raangu,
thanks for it, for the moment i think the most credits goes to martO,
who is compiling a lot for the moment, as we notice

You are always welcome to help us, the more help, the faster it goes
If you are looking to do something, there are a lot of unaswered questions in the topic, feel free to make some research onto it.

Update on my previous post;
i was mixing arm-linux with haret, my apologies for this mistakes.
(sometimes it is hard to follow )

The needed "Linux kernel machine type" can be found in the Haret CVS logs.
Most updated version you can find here :
CVS log for linux/kernel26/arch/arm/tools/mach-types

According revision 1.47 it provides following info :

# machine_is_xxx CONFIG_xxxx MACH_TYPE_xxx number
smdk6410 MACH_SMDK6410 SMDK6410 1626

so the default should be the following (to boot Qi):

Code:

set MTYPE     SMDK6410
set KERNEL start_qi_all-s3c6410
boot

or

Code:

 set MTYPE     1626
set KERNEL start_qi_all-s3c6410
boot

The information on haret is not clear to me for the moment if we should use "Mach_type_xxx" or "number". The log will say ok or failure if entry is not valid( i guess).

So give it a go and provide feedback

ps; for die hards with free time : you could try to run the patched haret version in combination with arm-linux aswel just to see wath happens

MartO, if you managed to compile a qi for s3c6410, could you provide a link aswel? Maybe some m8 users would like to test ..



ps yet another intersting link (about arm-linux and bootloaders )
http://www.arm.linux.org.uk/developer/booting.php

ooh, I didn't see the howtos and so on in your first posts... very nice...
I also have to thank MartO, you are right, forgot to write it

maybe this afternoon or tomorrow i will give debian in a vm a try and try to compile android on my own, i think this should be a good start into this whole thing

if the howto works, this shouldn't be a problem...

i will report about my experience an maybe ( if I have a good day and it works ) i will also do something like a long howto for compiling from scratch, with installing debian and so on... maybe some more interested will do some research...

Great, your my hero!
do not forget that the howtos in the topic are not always complete,
so do a bit searching on google to , just to make sure

so if you get there write it down step by step.
As tip i would say try to make a standard built first to get to know the commands and then try for the s3c6410

(and if you have spare time maybe you could get a look on the initrd file)

ok,
it will be a shame..
but..
i am going to quit the project...


lol, it is a joke what is was telling , i´m not going to quit this project, i have spended allready to much time for it, lol, but some humor looked nice ;-)

Now to be honest i got some pretty good news;
Like you know meizu try´s to develop android but is not really confirmed if it will come.

So thats why we try it ourself and everyone is invited to help with us, even the smallest help is appreciated.

And maybe, maybe life will get better and we could expect a whole lot of help in the future.. why?
read this url :
the::unwired - PHOTOVIEW: Samsung Omnia II Windows Mobile Smartphone

i will quote the interesting part :P

.... Running on an Samsung ARM 1176 based S3C6410 mobile application processor at 667 MHz, the Samsung Omnia II features quadband GSM/GPRS/EDGE a....
The Samsung Omnia II is expected to be released anytime end of August/early September

yep, you have readed my mind, if this launches we have a bunch of people who presumably will try to develop android for the same cpu like our m8.

on the otherhand i could be dreaming aloud:-)

back on topic any testing m8 owner with some news on performed tests?


update:
good news can make a good day,
found interesting links:
http://support-köhler.de/andromnia/viewtopic.php?f=2&t=5&sid=052f6024b7f6fdf613f2045a 3ac45b60
quote :

first you have to build the kernel
http://support-koehler.de/andromnia/vie ... 65e66ff852

next step is to build the busy box
http://support-koehler.de/andromnia/vie ... 65e66ff852

then we have to create the initrd file
http://support-koehler.de/andromnia/vie ... 65e66ff852

now we can configure haret and running the kernel
http://support-koehler.de/andromnia/vie ... 65e66ff852

Android on i900? - WinMo @ MoDaCo

Android - Porting Guide

===== HaRET 0.5.1 =====
Setting KMode to true.
Old KMode was 0
Finished initializing output
Loading dynamically bound functions
Unable to load library 'gx'
Unable to load library 'gx'
Unable to load library 'gx'
Unable to load library 'gx'
Function 'LoadLibraryExW' in library 'coredll' at 4003B480
Function 'GetSystemPowerStatusEx2' in library 'coredll' at 4002C86C
Function 'SleepTillTick' in library 'coredll' at 40026C40
Function 'AllocPhysMem' in library 'coredll' at 40026BD0
Function 'FreePhysMem' in library 'coredll' at 40026C14
Function 'CreateToolhelp32Snapshot' in library 'toolhelp' at 4011107C
Function 'Process32First' in library 'toolhelp' at 40111190
Function 'Process32Next' in library 'toolhelp' at 40111208
Function 'Module32First' in library 'toolhelp' at 401113EC
Function 'Module32Next' in library 'toolhelp' at 40111470
Function 'CloseToolhelp32Snapshot' in library 'toolhelp' at 401110CC
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Unable to load library 'ace_ddi'
Detecting memory
WinCE reports memory size 201326592 (phys=180318208 store=72040448)
Detecting current machine
Trying to detect machine (Plat='miniOne' OEM='Meizu miniOne')
Wince reports processor: core=ARM1176JZF-S name=S3C6410 cat= vend=Samsung Electronics
Looking at machine Alpine
Looking at machine Apache
Looking at machine AximX50
Looking at machine AximX5
Looking at machine Beetles
Looking at machine Blueangel
Looking at machine Himalaya
Looking at machine Magician
Looking at machine Universal
Looking at machine H1910
Looking at machine H1940
Looking at machine H2200
Looking at machine H3600b
Looking at machine H3700
Looking at machine H3870
Looking at machine H3900
Looking at machine H4000
Looking at machine H4300
Looking at machine H5000
Looking at machine H6340
Looking at machine HX2000
Looking at machine HX4700
Looking at machine Sable
Looking at machine Wizard
Looking at machine Hermes
Looking at machine Trinity
Looking at machine Athena
Looking at machine G500
Looking at machine Artemis
Looking at machine Herald
Looking at machine Prophet
Looking at machine RX3000
Looking at machine Treo700wx
Looking at machine e310
Looking at machine e740
Looking at machine Acer_n30
Looking at machine Mio_P550
Looking at machine Kaiser
Looking at machine Loox5xx
Looking at machine Jornada820
Looking at machine H3100
Looking at machine H3600a
Looking at machine Tornado
Looking at machine Libra
Looking at machine Generic Intel PXA27x
Looking at machine Generic Intel PXA
Looking at machine Generic TI OMAP
Looking at machine Generic TI OMAP15xx
Looking at machine Generic Samsung s3c24xx
Looking at machine Generic MSM7500
Looking at machine Generic ARM 920t
Looking at machine Generic ARM 926
Looking at machine Generic ARM v6
Looking at arch Generic Intel PXA27x
Exception on arch Generic Intel PXA27x detect
Looking at arch Generic Intel PXA
Exception on arch Generic Intel PXA detect
Looking at arch Generic TI OMAP
Exception on arch Generic TI OMAP detect
Looking at arch Generic TI OMAP15xx
Looking at arch Generic Samsung s3c24xx
Exception on arch Generic Samsung s3c24xx detect
Looking at arch Generic MSM7500
Exception on arch Generic MSM7500 detect
Looking at arch Generic ARM 920t
Exception on arch Generic ARM 920t detect
Looking at arch Generic ARM 926
Exception on arch Generic ARM 926 detect
Looking at arch Generic ARM v6
Exception on arch Generic ARM v6 detect
Not registering command IGPIO
Not registering command WG|PIO
Not registering command GPLR
Not registering command GPDR
Not registering command GAFR
Not registering command GPIO
Not registering command GPIOST
Registering command LOADLIBRARYEX
Registering command TRACES
Registering command RESUMETRACES
Registering command WI|RQ
Not registering command TRACE
Not registering command TRACEMASK
Not registering command TRACE2
Not registering command TRACETYPE
Not registering command TRACE2TYPE
Registering command TRACEFORWATCH
Not registering command INSN
Not registering command INSNREENABLE
Not registering command INSNREG1
Not registering command INSNREG2
Not registering command INSN2
Not registering command INSN2REENABLE
Not registering command INSN2REG1
Not registering command INSN2REG2
Registering command ALTL1TRACE
Registering command MAXL1TRACE
Registering command MAXL1TRACERESUME
Registering command MMUTRACE
Registering command TRACEIGNORE
Registering command PERMISSIVEMMUTRACE
Registering command KILL
Registering command PS
Registering command LSMOD
Registering command ADDR2MOD
Not registering command AC97
Not registering command ATIDBG
Initializing for machine 'Default'
Exception on cpu id detect
Welcome, this is HaRET 0.5.1 running on WindowsCE v6.0
Minimal virtual address: 00010000, maximal virtual address: 7FFFFFFF
Detected machine Default/generic (Plat='miniOne' OEM='Meizu miniOne')
CPU is ? running in user mode
Enter 'HELP' for a short command summary.
Running WSAStartup
Starting gui
In initdialog
Found machine Default
executing startup.txt
HaRET(1)# set mtype 1910
HaRET(2)# set kernel zImage
HaRET(3)# set initrd root.cpiogz
HaRET(4)# set ramsize 0x6000000
HaRET(5)# set ramaddr 0x10000000
HaRET(6)# set cmdline "root=/dev/ram0 init=/init console=tty0 mem=76M msmsdcc_id=3 imgdevname=/dev/mmcblk0p1 imgdevnum=1 imgdir=/tmp msmts_calib=0x6D.0x5D.0x340.0x375 msmvkeyb_toggle=hide"
HaRET(7)# boot
boot KERNEL=zImage INITRD=root.cpiogz
Opening file zImage
Opening file root.cpiogz
boot params: RAMADDR=10000000 RAMSIZE=06000000 MTYPE=1910 CMDLINE='root=/dev/ram0 init=/init console=tty0 mem=76M msmsdcc_id=3 imgdevname=/dev/mmcblk0p1 imgdevnum=1 imgdir=/tmp msmts_calib=0x6D.0x5D.0x340.0x375 msmvkeyb_toggle=hide'
Boot FB feedback: 1
Failed to lock 610 pages (code 6)
Calling WSACleanup
Shutting down

.............

isthisanynews?

hi MaikelRunia,

thanks for your post!
and ofcourse it is, i notice that the log says:

Code:

HaRET(1)# set mtype 1910

Could you inform us with what haret version/default.txt you have used?

tx ;-)

ps,
yet another usefull link :
http://alone-in-the-light.zenvoid.or...m-sd-card.html
=>

Index of zenvoid.org/software/qi-smartq/

linux-SMDK6410.bin
linux-kernel-src-2.6.24.7-smartq.tar.lzma
linux-modules-SMDK6410.tar.bz
qi-smartq-20090603.bin